R Notebook
Executive Summary
Target Audience
- Potential & Existing host
Purpose
- From our analysis, to find out how to be competitive during this time (political unrest, covid) for diff room types
Making sure all the correct libraries and installations are complete
Importing our data
Initial Data Exploration
Filtering our sub-dataframes
# Store original str price to a new column
listings$price_str = listings$price
# Clean listing price to numeric
listings$price <- as.numeric(gsub('[$,]', '', listings$price))
# Filter listings to only those <HKD5000
listings <-listings[!(listings$price > 5000 ),]Creating our sub-dataframes
Outputting the subdataframes as and when required
## [1] "Apartment" "Guest suite" "Condominium"
## [4] "Bed and breakfast" "Hostel" "Hotel"
## [7] "House" "Campsite" "Aparthotel"
## [10] "Hut" "Loft" "Guesthouse"
## [13] "Serviced apartment" "Boutique hotel" "Townhouse"
## [16] "Other" "Bungalow" "Tiny house"
## [19] "Cave" "Boat" "Pension (South Korea)"
## [22] "Igloo" "Chalet" "Cottage"
## [25] "Treehouse" "Nature lodge" "Tent"
## [28] "Cabin" "Villa" "Vacation home"
## [31] "Minsu (Taiwan)" "Island" "Earth house"
## [34] "Dorm" "Houseboat" "Farm stay"
## [37] "Castle" "Casa particular (Cuba)" "Kezhan (China)"
## [40] "Resort"Spatial Data Analysis
Reviews <- listings %>% group_by(neighbourhood_cleansed) %>% summarise(avg_loc_review = mean(review_scores_location, na.rm = TRUE))
colnames(Reviews) <- c("Eng_name","value")
Reviews$Eng_name = replace(Reviews$Eng_name, Reviews$Eng_name == "Central & Western", "Central and Western" )
library(sp)
hkmap = readRDS("HKG_adm1_old.rds") # geo data of HK map
# Preprocessing
map_data = data.frame(id=hkmap$ID_1, Eng_name=hkmap$NAME_1)
map_data = merge(map_data, Reviews, by = 'Eng_name')
hkmapdf = fortify(hkmap)
map_data = merge(hkmapdf, map_data, by="id")
map_bg = ggplot(map_data, aes(long, lat, group=group, fill = value)) +
geom_polygon() + # Shape
scale_fill_gradient(limits=range(map_data$value),
low="#9999FF", high="#330066") +
layer(geom="path", stat="identity", position="identity",
mapping=aes(x=long, y=lat, group=group,
color=I('#FFFFFF'))) +
guides(fill = guide_legend(title = "Rating", title.position = "top")) +
ggtitle("Map showing Average Location Score by Area")
map_bg = map_bg + theme(axis.line=element_blank(),
axis.text.x=element_blank(), axis.title.x=element_blank(),
axis.text.y=element_blank(), axis.title.y=element_blank(),
axis.ticks=element_blank(), panel.background = element_blank()
)
map_bg
Prices <- listings %>% group_by(neighbourhood_cleansed) %>% summarise(avg_price = mean(price, na.rm = TRUE))
colnames(Prices) <- c("Eng_name","value")
Prices$Eng_name = replace(Prices$Eng_name, Prices$Eng_name == "Central & Western", "Central and Western" )
library(sp)
hkmap2 = readRDS("HKG_adm1_old.rds") # geo data of HK map
# Preprocessing
map_data2 = data.frame(id=hkmap2$ID_1, Eng_name=hkmap2$NAME_1)
map_data2 = merge(map_data2, Prices, by = 'Eng_name')
hkmapdf2 = fortify(hkmap2)
map_data2 = merge(hkmapdf2, map_data2, by="id")
map_bg = ggplot(map_data2, aes(long, lat, group=group, fill = value)) +
geom_polygon() + # Shape
scale_fill_gradient(limits=range(map_data2$value),
low="#FF9999", high="#CC0000") +
layer(geom="path", stat="identity", position="identity",
mapping=aes(x=long, y=lat, group=group,
color=I('#FFFFFF'))) +
guides(fill = guide_legend(title = "Rating", title.position = "top")) +
ggtitle("Map showing Average Price by Area")
map_bg = map_bg + theme(axis.line=element_blank(),
axis.text.x=element_blank(), axis.title.x=element_blank(),
axis.text.y=element_blank(), axis.title.y=element_blank(),
axis.ticks=element_blank(), panel.background = element_blank()
)
map_bg
### Price per person scatter plots
Pricesperperson <- select(listings,'accommodates', 'price','review_scores_rating','neighbourhood_cleansed')
Pricesperperson <- within(Pricesperperson, price_per_pax <- price/accommodates)
Pricesperperson<-Pricesperperson[!(Pricesperperson$price_per_pax > 10000 || Pricesperperson$price_per_pax < 20 ),]
ggplot(data = Pricesperperson, mapping = aes(x = price_per_pax, y = review_scores_rating, color = neighbourhood_cleansed)) + geom_point()## Warning: Removed 4291 rows containing missing values (geom_point).
Listings per neighbourhood
propertydf <- listings %>% group_by(neighbourhood_cleansed, property_type) %>% summarize(Freq = n())
propertydf <- propertydf %>% filter(property_type %in% c("Apartment","House","Condominium","Townhouse", "Loft"))
totalproperty<- listings %>% filter(property_type %in% c("Apartment","House","Condominium","Townhouse", "Loft"))%>% group_by(neighbourhood_cleansed) %>% summarize(sum = n())
propertyratio <- merge(propertydf, totalproperty, by="neighbourhood_cleansed")
propertyratio <- propertyratio %>% mutate(ratio = Freq/sum)
ggplot(propertyratio, aes(x=neighbourhood_cleansed, y=ratio, fill = property_type)) +
geom_bar(position = "stack",stat="identity") + xlab("Borough") + ylab("Count")+
scale_fill_discrete(name = "Property Type") +
scale_y_continuous(labels = scales::percent) +
ggtitle("Which types of Listings are there in Hong Kong?",
subtitle = "Map showing Count of Listing Type by Borough ") +
theme(plot.title = element_text(face = "bold")) +
theme(plot.subtitle = element_text(face = "bold", color = "grey35")) +
theme(plot.caption = element_text(size=10, color = "grey68"))+scale_color_gradient(low="#d3cbcb", high="#852eaa")+
theme(axis.text.x = element_text(size = 8) ) +
scale_fill_manual("Property Type", values=c("#FF3300","#F0E442", "#009E73", "#56B4E9", "#666666")) +
xlab("Neighborhood") + ylab("Percentage")+coord_flip()
Price Comparisons
Demand and Price Analysis
### Demand Price Analysis 
Sentitiment Analysis via Review WordCloud
